JP2018025949A - Learning device, image search device, method, and program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a learning device capable of learning a ranking model for accurately searching an image appropriate for a text data of a music.SOLUTION: A ranker learning unit 64 learns a ranking model based on a keyword pair feature quantity of a correct solution, a keyword pair feature quantity of an incorrect solution, a topic pair feature quantity of the correct solution, and a topic pair feature quantity of the incorrect solution to thereby learn a ranking model for accurately searching an image which is appropriate for a text data of a music.SELECTED DRAWING: Figure 1

Description

  The present invention relates to a learning device, a video search device, a method, and a program, and more particularly, to a learning device, a video search device, a method, and a program for searching for a video suitable for song lyrics data.

  There is a great need to search video from text, such as image search in Web search. If the video can be searched from the text, it is not necessary to search while visually checking the video, and the cost can be reduced. In addition, if an image suitable for the text can be acquired, it is possible to visually assist the content of the text.

Information Retrieval and Language Processing (Language and Computation), Chapter 2 Information Retrieval Fundamentals, Chapter 4 Use of Language Processing Technology, Takenobu Tokunaga (Author), The University of Tokyo Press, 1999

  In an image search or the like, the input text is often a keyword. However, there has not been established a technique for searching for a corresponding video by inputting a long and subjective sentence such as a song lyrics.

  The present invention has been made in view of the above circumstances, and provides a learning device, method, and program capable of learning a ranking model for accurately searching for a video suitable for song lyrics data. With the goal.

  It is another object of the present invention to provide a video search apparatus, method, and program capable of accurately searching for a video suitable for song lyrics data.

  In order to achieve the above object, a learning device according to a first aspect of the present invention is a learning device for learning a ranking model for searching for a video suitable for song lyrics data, wherein the song lyrics data and the song The correct answer data that is a pair with the summary text data given to the video suitable for the lyric data, and the lyric data of the music, and the summary text data given to the video that is not suitable for the lyric data of the music A keyword extraction unit for lyrics that extracts lyrics keywords from each of lyrics data included in certain incorrect answer data, and a summary text keyword is extracted from each of correct text and summary text data included in the incorrect answer data. A keyword extractor for summary text, a keyword for lyrics extracted from the lyrics data of the correct answer data, and the correct answer Keyword pair feature amount representing a combination of keywords for summary text extracted from the summary text data of the data, keywords for lyrics extracted from the lyrics data of the incorrect answer data, and the summary text data of the incorrect answer data And a ranker learning unit that learns the ranking model based on the keyword pair feature amount representing the combination of the keywords for the summary text extracted from.

  The learning device according to the second invention is a learning device for learning a ranking model for searching for a video suitable for song lyrics data, and is suitable for song lyrics data and song lyrics data. Included in correct answer data that is a pair with summary text data given to video and incorrect answer data that is a pair of lyrics data of music and summary text data given to video that is not suitable for the lyrics data of the music A topic extraction unit for lyrics that extracts a topic for lyrics from each of the lyrics data to be extracted, and a topic extraction for summary text that extracts a topic for summary text from each of the summary text data included in the correct answer data and the incorrect answer data Section, the topic for lyrics extracted from the lyrics data of the correct answer data, and the summary text data of the correct answer data A topic pair feature amount representing a combination of topics extracted from the summary text, a topic for lyrics extracted from the lyrics data of the incorrect answer data, and a summary text extracted from the summary text data of the incorrect answer data A ranker learning unit that learns the ranking model based on a topic pair feature amount representing a combination of topics.

  A learning device according to a third aspect of the present invention is a learning device for learning a ranking model for searching for a video suitable for song lyrics data, and is suitable for song lyrics data and the song lyrics data. Included in correct answer data that is a pair with summary text data given to video and incorrect answer data that is a pair of lyrics data of music and summary text data given to video that is not suitable for the lyrics data of the music Lyric keyword extraction unit for extracting lyric keywords from each of the lyric data to be extracted, and general text keyword extraction for extracting the general text keywords from each of the general text data included in the correct data and the incorrect data And a song that extracts a lyric topic from each of the correct answer data and the lyric data included in the incorrect answer data A topic extraction unit for extracting a summary text topic from each of the summary text data included in the correct answer data and the incorrect answer data, and extracted from the lyrics data of the correct answer data. A keyword pair feature amount representing a combination of a keyword for lyrics and a keyword for summary text extracted from the summary text data of the correct answer data, a keyword for lyrics extracted from the lyrics data of the incorrect answer data, and the incorrect answer data A keyword pair feature amount representing a combination of keywords for summary text extracted from the summary text data, a topic for lyrics extracted from the lyrics data of the correct answer data, and the summary text data of the correct answer data Of the topic for summary text A topic pair feature representing a combination of a topic pair feature amount representing matching, a topic for lyrics extracted from the lyrics data of the incorrect answer data, and a topic for summary text extracted from the summary text data of the incorrect answer data And a ranker learning unit that learns the ranking model based on the quantity.

  In the learning device according to the first to third inventions, the summary text keyword may be a keyword representing a person, a place, a season, or an event.

  A video search device according to a fourth invention is a video search device for searching for a video suitable for lyric data of a song, the lyric keyword extracting unit extracting a lyric keyword from the input lyric data of the song, The lyric keywords extracted from the lyric data of correct answer data that is a pair of the lyric data of music and the outline text data assigned to the video suitable for the lyric data of the music, and the outline text data of the correct data Of incorrect answer data that is a pair of a keyword pair feature amount representing a combination of keywords for summary text extracted from the above, lyrics data of music, and summary text data attached to a video not suitable for the lyrics data of the music Extracted from the lyric keyword extracted from the lyric data and the summary text data of the incorrect answer data A ranking model storage unit for storing a ranking model for searching for a video suitable for the lyrics data of a song, learned in advance based on a keyword pair feature amount representing a combination of keywords for the summary text, and for each of the videos , Based on a keyword pair feature amount representing a combination of a keyword for lyrics extracted by the keyword extraction unit for lyrics and a keyword for summary text extracted from the summary text data attached to the video, and the ranking model, A video search unit that searches for a video suitable for the lyrics data of the input music.

  According to a fifth aspect of the present invention, there is provided a video search apparatus for searching for a video suitable for lyric data of a song, wherein the lyric topic extraction extracts a lyric topic from the input lyric data of the music. Section, the lyrics topic extracted from the lyrics data of the correct answer data that is a pair of the lyrics data of the song, and the summary text data attached to the video suitable for the lyrics data of the song, and the overview of the correct answer data Incorrect answer that is a pair of topic pair feature amount representing a combination of topics for summary text extracted from text data, lyrics data of music, and summary text data attached to video not suitable for the lyrics data of the music Topic for lyrics extracted from the lyrics data of the data and the summary extracted from the summary text data of the incorrect answer data A ranking model storage unit for storing a ranking model for searching for a video suitable for the lyrics data of the music, which has been learned in advance based on a topic pair feature amount representing a combination of topics for text, and for each of the videos, Based on a topic pair feature amount representing a combination of a topic for lyrics extracted by the topic extraction unit for lyrics and a topic for summary text extracted from the summary text data attached to the video, and an input based on the ranking model And a video search unit for searching for a video suitable for the lyrics data of the recorded music.

  According to a sixth aspect of the present invention, there is provided a video search device for searching for a video suitable for lyric data of a song, wherein the lyric keyword is extracted from the input lyric data of the tune. A lyric topic extracting unit for extracting a lyric topic from the input lyric data, lyric data of the tune, and summary text data attached to a video suitable for the lyric data of the tune Keyword pair feature amount representing a combination of a keyword for lyrics extracted from the lyrics data of the correct answer data as a pair and a keyword for summary text extracted from the summary text data of the correct answer data, lyrics data of music, and The lyrics data of incorrect answer data that is a pair with the summary text data attached to the video not suitable for the lyrics data of the music A keyword pair feature amount representing a combination of a keyword for lyrics extracted from the data and a keyword for summary text extracted from the summary text data of the incorrect answer data, and a topic for lyrics extracted from the lyrics data of the correct data And a topic pair feature amount representing a combination of topics for summary text extracted from the summary text data of the correct answer data, and a topic for lyrics extracted from the lyrics data of the incorrect answer data and the summary of the incorrect answer data A ranking model storage unit for storing a ranking model for searching for videos suitable for lyric data of a song, learned in advance based on a topic pair feature amount representing a combination of topics for summary text extracted from text data; The lyrics for each of the videos A keyword pair feature amount representing a combination of a keyword for lyrics extracted by the keyword extraction unit and a keyword for summary text extracted from the summary text data attached to the video, and for the lyrics extracted by the topic extraction unit for lyrics Search for a video suitable for the lyrics data of the input music based on the topic pair feature amount representing the combination of the topic for the summary text extracted from the topic and the summary text data attached to the video and the ranking model. And a video search unit.

  A learning method according to a seventh aspect of the present invention is a learning method in a learning device for learning a ranking model for searching for a video suitable for song lyrics data, wherein the lyrics keyword extraction unit includes song lyrics data, Correct answer data that is a pair with summary text data given to the video suitable for the lyrics data of the music, and the lyrics data of the music, and summary text data given to the video not suitable for the lyrics data of the music A step of extracting a lyric keyword from each of lyric data included in a pair of incorrect answer data, and a summary text keyword extracting unit, from each of the correct text and the summary text data included in the incorrect answer data, A step of extracting a keyword for summary text and a ranker learning unit are extracted from the lyrics data of the correct answer data. A keyword pair feature amount representing a combination of a keyword for lyrics and a keyword for summary text extracted from the summary text data of the correct answer data, a keyword for lyrics extracted from the lyrics data of the incorrect answer data, and the incorrect answer data Learning the ranking model based on a keyword pair feature amount representing a combination of keywords for summary text extracted from the summary text data.

  According to an eighth aspect of the present invention, there is provided a video search method comprising: a keyword extraction unit for lyrics; lyrics data of music; and incorrect answer data that is a pair of summary text data attached to video not suitable for the lyrics data of the music To the lyrics data of the music learned in advance based on the keyword pair feature amount representing the combination of the keyword for lyrics extracted from the lyrics data and the keyword for summary text extracted from the summary text data of the incorrect answer data A video search method in a video search apparatus, comprising: a ranking model storage unit for storing a ranking model for searching for a suitable video; and a video search unit, wherein the video search device searches for a video suitable for lyric data of music. A keyword extracting unit for extracting lyrics keywords from the lyrics data of the input music; A keyword pair feature amount representing a combination of a lyric keyword extracted by the lyric keyword extracting unit and an outline text keyword extracted from the outline text data attached to the image for each of the videos by the image search unit Lyric data extracted from the lyric data of the correct answer data and the summary text of the correct answer data, which is a pair of the lyric data of the tune and the outline text data given to the video suitable for the lyric data of the tune Based on a keyword pair feature amount representing a combination of keywords for summary text extracted from data and a ranking model stored in the ranking model storage unit, a video suitable for the lyrics data of the input music is searched. And executing the steps.

  A program according to a ninth invention is a program for causing a computer to function as each unit of the learning device according to the first to third inventions or the video search device according to the fourth to sixth inventions.

  According to the learning device, method, and program of the present invention, a keyword for lyrics is extracted from each of lyrics data, a keyword for summary text is extracted from each of summary text data, and the lyrics are extracted from each of the lyrics data. A topic is extracted, a topic for summary text is extracted from each of the summary text data, and a combination of a keyword for lyrics extracted from the lyrics data of correct answer data and a keyword for summary text extracted from the summary text data of correct data Keyword pair feature amount that represents the combination of the keyword for lyrics extracted from the lyrics data of the incorrect answer data and the keyword for summary text extracted from the summary text data of the incorrect answer data, and the lyrics of the correct answer data Lyric topics and correct answer data extracted from the data Topic pair features that represent a combination of topics for summary text extracted from summary text data, and topics for lyrics extracted from lyrics data of incorrect answer data and summary text extracted from summary text data of incorrect answer data By learning the ranking model based on the topic pair feature amount that represents the combination of topics, it is possible to learn the ranking model for accurately searching the video suitable for the lyrics data of the music. It is done.

  According to the video search apparatus, method, and program of the present invention, the lyrics keyword is extracted from the lyrics data of the input music, the topic for lyrics is extracted from the lyrics data of the input music, and each of the videos , The keyword pair feature amount representing the combination of the keyword for summary text extracted from the extracted keyword for lyrics and the summary text data attached to the video, and the summary text data attached to the extracted topic for lyrics and video Video suitable for the lyrics data of the music by searching the video suitable for the lyrics data of the input music based on the topic pair feature amount representing the combination of topics for the summary text extracted from the ranking model Can be searched with high accuracy.

It is a block diagram which shows the structure of the learning apparatus which concerns on embodiment of this invention. It is a figure which shows an example of the result of the morphological analysis of lyrics data. It is a figure which shows an example of the result of word extraction. It is a figure which shows an example of the result of the morphological analysis of an outline text. It is a figure which shows an example of the top five probability values of the topic of a certain lyrics. It is a block diagram which shows the structure of the video search apparatus which concerns on embodiment of this invention. It is a figure which shows an example of the ranking result of an outline text. It is a flowchart which shows the learning process routine in the learning apparatus which concerns on embodiment of this invention. It is a flowchart which shows the video search process routine in the video search device which concerns on embodiment of this invention.

  Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.

<Outline according to Embodiment of the Present Invention>

  First, an outline of the embodiment of the present invention will be described.

  The music video search in the embodiment of the present invention includes a learning process and a search process. In the learning process, a ranking model necessary for performing the search process is created in the learning device. In the search process, the video search device searches the lyric data of the music based on the ranking model created in the learning process.

<Configuration of Learning Device According to Embodiment of the Present Invention>

  Next, the configuration of the learning device according to the embodiment of the present invention will be described. As shown in FIG. 1, a learning device 100 according to an embodiment of the present invention is a computer that includes a CPU, a RAM, and a ROM that stores a program for executing a learning processing routine described later and various data. Can be configured. The learning device 100 functionally includes a calculation unit 20 as shown in FIG.

  The calculation unit 20 includes a lyrics data DB 30, a keyword extraction unit 32 for lyrics, a keyword list 34 for each lyrics, a summary text data DB 36, a keyword extraction unit 38 for summary text, a keyword list 40 for each summary text, Topic model creation unit 42 for lyrics, topic model 44 for lyrics, topic extraction unit 46 for lyrics, topic list 48 for each lyrics, topic model creation unit 50 for summary text, topic model 52 for summary text, Topic extraction section 54 for summary text, topic list 56 for each summary text, lyrics / summary text pair correct answer data 58, incorrect answer data creation section 60, lyrics / summary text pair incorrect answer data 62, ranker learning section 64 and a ranking model storage unit 66.

  The lyrics data DB 30 stores lyrics data of a plurality of songs.

  The lyric keyword extracting unit 32 extracts lyric keywords from each of the lyric data stored in the lyric data DB 30 as described below, creates a keyword list for each lyric, Save as list 34.

  Specifically, first, the lyric keyword extracting unit 32 performs morphological analysis for each of the lyrics in the lyric data.

  For example, suppose you have the following lyrics:

I want to see you soon

  The above lyrics are morphologically analyzed as shown in FIG. Here, JTAG developed by NTT (R) is used for the morphological analyzer.

  In FIG. 2, each line represents one word, and from the left, the surface form, the part of speech, the standard form, the basic form, the reading, and the semantic attribute are represented. The semantic attribute is composed of three fields: a semantic attribute related to a noun, a semantic attribute related to a proper noun, and a semantic attribute related to a predicate. A semantic attribute is a number representing meaning.

  From the above morphological analysis results, it is understood that “you” has 15 and 2651 as semantic attributes. Details of the semantic attributes are shown in Non-Patent Document 2 below.

Non-patent document 2: Satoru Ikehara, Masahiro Miyazaki, Satoshi Shirai, Akio Yokoo, Hiroaki Nakaiwa, Kentaro Ogura, Yoshifumi Oyama, Yoshihiko Hayashi (1997) Iwanami Shoten.

  Next, the lyric keyword extracting unit 32 extracts a predetermined word as a keyword by using the result of the morphological analysis. Specifically, a word whose part of speech is any of a noun, a verb stem, and an adjective stem and is not a suffix is extracted. As a result of the above morphological analysis, the words shown in FIG. 3 are extracted.

  Here, basic forms are extracted from morphemes whose part of speech is a verb stem and an adjective stem. The standard form is extracted from the morpheme whose part of speech is a noun.

  As a result of the above procedure, the following keyword list is created for the target lyrics.

[Me, you, meet, meet now ...]

  The keyword extraction process is applied to all lyrics in the lyrics data, a keyword list is created for each lyrics, and is stored as the keyword list 34 for each lyrics.

  The summary text data DB 36 stores summary text added to the video. The summary text may be anything as long as it is a description of the video, a script of the video, conversation information in the video, or text. In this embodiment, the text in which the video description is written is referred to as summary text.

  As will be described below, the summary text keyword extraction unit 38 extracts summary text keywords from each of the summary text data stored in the summary text data DB 36 and creates a keyword list for each summary text. The keyword list 40 of each summary text is stored. The keyword list 40 for each summary text is a list of keywords representing the situation in the video described in the summary text.

  Specifically, the summary text keywords include the following six types of keywords.

・ Person keyword: Words and expressions representing the person in the image ・ Location keyword: Words and expressions representing the place in the image ・ Season keyword: Words and expressions representing the season in the image ・ Event keyword: Words representing the event in the image Expression / behavioral keywords: Words and expressions / emotions that represent actions and actions in the video Keywords: Words and expressions that represent emotions in the video

  Other keywords may be defined depending on the application. Here, keywords that are considered important for video retrieval are defined.

  The summary text keyword extracting unit 38 uses a Japanese vocabulary system (not shown), an emotion word extractor (not shown), and an evaluation expression extractor (not shown) to extract these keywords.

  The Japanese vocabulary system is used to extract words representing person keywords, place keywords, seasonal keywords, and event keywords from the above situation. In the Japanese vocabulary system, semantic attributes of nouns are arranged hierarchically, and each semantic attribute has an upper semantic attribute and a lower semantic attribute.

  For example, “specific” is above the semantic attribute of “location”, and “facility”, “region”, and “nature” are below. Using the semantic attribute hierarchy information, the keywords considered to represent each situation are listed as follows.

・ Person keywords: “person”, “clothes”, “clothing”, and words / place keywords corresponding to semantic attributes below these: “place”, “building”, “vehicle”, “workplace”, and , Words / season keywords corresponding to these lower semantic attributes: “season”, and words / event keywords corresponding to these lower semantic attributes: “expression / event”, “life”, and these Words corresponding to lower semantic attributes

  Here, if there are words in the summary text that match the listed keywords, they are extracted as persons, places, seasons, or event keywords.

  For example, consider the following summary text:

Summer boy goes to the park

  FIG. 4 shows the result of the morphological analysis performed on the above summary text. As described above, semantic attributes are described in the last column.

  Here, “summer” has a semantic attribute of 2674 (summer) and is a subordinate attribute of 2672 (season), and thus is extracted as a seasonal keyword. Currently, when a word has a plurality of semantic attributes, extraction is performed using the first semantic attribute, but all may be used.

  When extracting a word, the standard form (third column) of the word is extracted. By using the standard form, keywords can be extracted while absorbing slight differences in notation.

  For the above summary text, the following words are extracted:

Character keyword "boy"
Location keyword "park"
Seasonal keyword “summer”
No event keywords

  Next, the emotion word extractor will be described. The emotion word extractor is used to extract emotion keywords. As the emotion word extractor, a language processing tool called rich indexer developed by NTT (R) is used. The rich indexer has a function of extracting a keyword related to emotion based on a predetermined keyword list. Using this rich indexer function, for example, keywords such as fun, surprise, sad, happiness, relief, and worry are extracted as emotion keywords.

  Next, the evaluation expression extractor will be described. The evaluation expression extractor is used to extract behavior / emotion keywords. Although there are various behaviors, the behaviors related to polarity such as positive and negative are regarded as important behaviors / emotions related to the video, and those related to the behaviors / emotions are extracted from the evaluation expression. Evaluation expression refers to language expression related to evaluation and emotion. Here, a list of evaluation expressions related to behaviour / emotion is created, and a list that matches them is extracted to obtain a behaviour / emotion keyword. For example, keywords such as smiling, slowly, refreshing, cutting, and panicking are extracted.

  The summary text keyword extraction unit 38 uses the above Japanese vocabulary system, emotion word extractor, and evaluation expression extractor to execute summary text for all summary text data stored in the summary text data DB 36. Keywords are extracted, a keyword list is created for each summary text, and saved as a keyword list for each summary text.

  The lyric topic model creation unit 42 creates a lyric topic model 44 from each of the lyric data stored in the lyric data DB 30.

  A topic model is a model that assumes that a document is generated from a plurality of potential topics, and is often used to analyze not only the surface layer of words but also the structure behind the document. The following non-patent document 3 is detailed regarding the topic model.

Non-Patent Document 3: Topic Model, Koji Iwata (Author), Kodansha, 2015.

  Specifically, the lyrics topic model creation unit 42 creates a topic model from the lyrics data using an algorithm called Latent Dirichlet Allocation (LDA). This is a common algorithm for creating topic models. To construct a topic model, each document (that is, each of lyric data) needs to be represented by a word set. Here, all words obtained as a result of morphological analysis are used. The number of topics was 300. A library called gensim was used as a topic model creation tool. About LDA, since there are many free software, you may use them.

  The lyrics topic extraction unit 46 extracts the lyrics topic from each of the lyrics data stored in the lyrics data DB 30 based on the lyrics topic model 44 and creates a topic list 48 of each lyrics. As described above, in the topic model, a potential topic existing behind a document is assumed, and they are mixed to generate a single document. In other words, by analyzing a single document using a topic model (this is called inference), the contained potential topics can be examined. Potential topics that are included in a document in large numbers are considered to be the main topics of the document, so they are extracted.

  When extracting topics for the lyrics data, the topic model 44 for lyrics is used. A topic list in the lyrics data is created by analysis using the topic model for lyrics. Specifically, in each lyric, only the topics included in a certain ratio or more are extracted and used as the topic list of the lyric. In the embodiment of the present invention, this threshold is set to 0.1. For example, when the degree of inclusion (probability value) of the top five topics of a certain lyric is as shown in FIG. 5, by extracting topics whose threshold is 0.1 or more, topics 85 and 122 are converted to the lyrics. Extract as a topic list. 85 and 122 are numbers representing topics.

  The summary text topic model creation unit 50 performs the same processing as that of the lyrics topic model creation unit 42 on the summary text data stored in the summary text data DB 36, so that the summary text data A topic model 52 is created. In the present embodiment, the number of topics is 50.

  The summary text topic extraction unit 54 extracts a summary text topic from each of the summary text data stored in the summary text data DB 36 based on the summary text topic model 52, and a topic list 56 of each summary text. Create The threshold value is the same as that of the lyrics topic extraction unit 46.

  The lyrics / outline text pair correct answer data 58 is a set of correct data pairs in which the lyrics data of music and the outline text data attached to the video suitable for the lyrics data of the music are correctly associated. These associations are performed manually.

  The incorrect answer data creating unit 60 uses the lyrics / summary text pair correct answer data 58 to create a pair of lyrics data corresponding to the song and summary text data attached to the video not suitable for the song lyrics data. The lyrics / summary text pair incorrect answer data 62 as a set is created.

  The purpose of this embodiment is to search for a video associated with the summary text by searching for the summary text that matches the lyrics. In other words, it is only necessary to classify lyrics that do not correspond to those that correspond from a plurality of summary texts.

  In general, classification problems are performed by supervised learning. For that purpose, both correct answer data (referred to as a positive example) and incorrect answer data (referred to as a negative example) are required.

  However, the teacher data has a positive example associated with it, but does not have a negative example, so the negative example is automatically generated. Specifically, the lyrics were randomly selected from a set of predetermined summary texts, and this was used as a negative example. Randomly chosen summary texts are rarely associated, and can be used as negative examples. Such a method is also called pseudo negative example generation, and is a method often used in machine learning.

  In this way, the incorrect answer data creating unit 60 creates a pair in which lyrics and summary text (summary text other than those associated with lyrics) are randomly combined as incorrect answer data. In addition to selecting randomly, a pair of lyrics and summary text that has been confirmed not to be handled manually may be used as incorrect answer data.

  In the present embodiment, the ratio of positive examples to negative examples is set to 1: 1. This ratio may be changed according to the performance of the ranker learning unit 64 in the subsequent stage.

  One incorrect answer data was created for each of the lyrics / outline text pair correct answer data 58 and used as learning data. The learning data further includes lyrics data and summary text associated with the lyrics, and lyrics data and summary text not associated with the lyrics data.

  The ranker learning unit 64 uses a keyword pair feature amount representing a combination of a keyword for lyrics extracted from lyrics data of correct answer data and a keyword for outline text extracted from summary text data of correct answer data, and lyrics data of incorrect answer data. Keyword pair feature amount representing a combination of keywords for summary text extracted from summary text data of extracted lyrics keywords and incorrect answer data, and topic text for lyrics and summary text of correct answer data extracted from lyrics data of correct data A combination of topic pair features that represent a combination of topics for summary text extracted from data, a topic for lyrics extracted from lyrics data of incorrect answer data, and a topic for summary text extracted from summary text data of incorrect answer data Topic representing Based on the A feature amount, you learn the ranking model, and stores the ranking model storage unit 66. In the above, the correct answer data is based on the lyrics / summary text pair correct answer data 58, and the incorrect answer data is based on the lyrics / summary text pair incorrect answer data 62.

  The ranker learning unit 64 extracts a feature amount from the learning data and learns an evaluation function based on the feature amount, whereby a ranking model (ranker) capable of ranking videos suitable for song lyrics data. ).

  The feature amount is extracted using a keyword list 34 for each lyrics and a keyword list 40 for each summary text, a topic list 48 for each lyrics, and a topic list 56 for each summary text obtained from each of the learning data. There are two types of feature quantities, including keyword pair feature quantities and topic pair feature quantities. The respective feature amounts are extracted for all combinations of the lyrics / summary text pair correct answer data 58 for correct answer data and the lyrics / summary text pair incorrect answer data 62 for incorrect answer data.

  The keyword pair feature amounts are the words included in the keyword list 34 of each lyric of the lyrics data created by the lyric keyword extracting unit 32 and the summary texts of the summary text created by the summary text keyword extracting unit 38. Based on the words included in the keyword list 40, all the combinations are used as feature amounts. For example, it is assumed that each of the keyword lists of lyrics data and summary text is configured as follows.

Keyword list of lyrics: [You (4 times), Meet (3 times), Pounding (1 time), ...]
Keyword list of summary text: [Casual (2 times), 2 people (2 times), Fun (1 time), ...]

  In this case, from the combination of the keyword list of the above lyrics and the keyword list of the summary text, the keyword pair feature amounts are “Kimi-Casual”, “Kimi-Two”, “Kimi-fun”, “Meet-Casual”. Extract features such as "Meeting-Two people", "Meeting-Fun", "Pounding-casual", "Pounding-two people", and "Pounding-fun". In this manner, keyword pair feature amounts are extracted for all combinations of the keyword list for lyrics and the keyword list for summary text.

  As the value of the keyword pair feature value, whether or not the keyword pair feature value appears in a combination having the keyword pair feature value, that is, a combination of the keyword list of lyrics and the keyword list of summary text is used as a binary value. Note that the frequency information in the combination may be used as the feature value instead of the binary value.

  The feature used for learning may be only a feature amount that appears in a certain number or more in all positive examples of the learning data. By doing so, the feature space is reduced and the learning cost is reduced. In the embodiment of the present invention, the threshold of the number of appearances is set to 5. That is, feature quantities included in the combination of learning data by 5 or more are used for learning.

  The topic pair feature amount is based on the topics included in the topic list 48 for each lyrics created by the topic extraction unit 46 for lyrics and the topic list 56 for each summary text created by the topic extraction unit 54 for summary text. Like the keyword pair feature quantity, all the combinations are made feature quantities. For example, it is assumed that the topic lists of the lyrics data and the summary text are configured as follows.

Lyrics topic list: [85,122]
Summary text topic list: [33,2,27]

  In this case, topic pair feature values “85-33”, “85-2”, “85-27”, “122-33”, “122-2”, “122-27” are extracted as feature values. Is done.

  Again, in all positive examples of learning data, only feature quantities that appear in a certain number or more may be used, but in the embodiment of the present invention, all topic pair feature quantities in the positive examples of learning data are used. Yes.

  In this way, keyword pair feature values and topic pair feature values are extracted for each of positive examples and negative examples in the learning data, and an evaluation function that can classify positive examples and negative examples is learned by machine learning. Since this is a simple binary classification problem, an algorithm often used for the classification problem may be used. Here, logistic regression is used. As another algorithm, a support vector machine (SVM) may be used. In addition, since the binary classification problem model can generally output the reliability of the likelihood of being positive (or the likelihood of being negative) for the case of the classification target, it is possible to rank a plurality of classification targets by using the numerical values. In the embodiment of the present invention, ranking is performed using a classification model obtained by logistic regression. Note that learning may be performed using a machine learning algorithm specialized for ranking, such as ranking SVM. In that case, the evaluation function may be learned so that positive examples are ranked higher than negative examples.

  The ranking model storage unit 66 stores a ranking model for searching for videos suitable for the lyrics data of the music learned by the ranker learning unit 64.

<Configuration of Video Retrieval Device According to Embodiment of the Present Invention>

  Next, the configuration of the video search apparatus according to the embodiment of the present invention will be described. As shown in FIG. 6, the video search apparatus 200 according to the embodiment of the present invention includes a CPU, a RAM, and a ROM that stores a program and various data for executing a video search processing routine described later. Can be configured with a computer. Functionally, the video search apparatus 200 includes an input unit 210, a calculation unit 220, and an output unit 270 as shown in FIG.

  The input unit 210 receives lyric data of music. In addition to the lyrics, information such as a singer name, a composer name, and a music genre may be received. The lyrics information is output to the calculation unit 220 for analysis.

  The calculation unit 220 includes a keyword extraction unit 232 for lyrics, a keyword list 240 for each summary text, a topic model for lyrics 244, a topic extraction unit for lyrics 246, a topic list 256 for each summary text, and a video search unit 264. And a ranking model storage unit 266.

  The lyric keyword extracting unit 232 extracts lyric keywords from the lyric data of the music received by the input unit 210. The specific process may be performed in the same manner as the keyword extraction unit 32 for lyrics.

  The keyword list 240 for each summary text stores the same items as the keyword list 40 for each summary text.

  The lyrics topic model 244 stores the same as the lyrics topic model 44.

  The lyrics topic extraction unit 246 extracts the lyrics topic from the lyrics data of the music received by the input unit 210 based on the lyrics topic model 244. The specific process may be the same process as the topic extraction unit 46 for lyrics.

  The ranking model storage unit 266 stores the same items as the ranking model storage unit 66.

  The video search unit 264 includes, for each video, a keyword for lyrics extracted by the keyword extracting unit for lyrics 232 and a summary text keyword extracted from the summary text data added to the video in the keyword list 240 for each summary. A combination of a keyword pair feature amount representing a combination, a lyric topic extracted by the lyric topic extraction unit 246, and a summary text topic extracted from the summary text data added to the video in the topic list 256 of each summary text Is searched for based on the topic pair feature quantity indicating the image and the ranking model stored in the ranking model storage unit 266.

  First, the video search unit 264 creates a pair of the lyric keyword list extracted by the lyric keyword extraction unit 232 and the keyword list 240 of the summary text to be searched, and extracts the above-described keyword pair feature amount. Further, a pair of the lyrics topic list extracted by the lyrics topic extraction unit 246 and the topic list 256 of the summary text to be searched is created, and the above-described topic pair feature amount is extracted. Then, by applying a ranking model (a logistic regression model in the embodiment of the present invention) stored in the ranking model storage unit 266 for each of the keyword pair feature amount and the topic pair feature amount, Find the score of positiveness (ie, whether it corresponds). By ranking the summary text based on this score, it is possible to obtain the summary text most likely to be associated, and as a result, it is possible to output a video associated with the summary text.

  For example, FIG. 7 shows the result of ranking the summary texts given to five videos for certain lyrics data.

  The result of FIG. 7 shows the number of the summary text and the output score (probability value indicating the likelihood of being a positive example because logistic regression is used). The higher this value, the more the lyrics correspond to the summary text Indicates that it is attached. In this example, the second summary text has a high score of 0.909, and it is determined that it corresponds most with the input lyrics data.

  The output unit 270 outputs N videos associated with the top N summary texts based on the summary text ranking information output by the video search unit 264. N may be set according to the application. In the embodiment of the present invention, N is set to 5. However, if it is desired to search only the most associated video, N may be set to 1.

<Operation of Learning Device According to Embodiment of the Present Invention>

  Next, the operation of the learning device 100 according to the embodiment of the present invention will be described. The learning device 100 executes a learning process routine shown in FIG.

  First, in step S100, a lyric keyword is extracted from each of the lyric data stored in the lyric data DB 30, a keyword list for each lyric is created, and saved as a keyword list 34 for each lyric.

  Next, in step S102, a summary text keyword is extracted from each of the summary text data stored in the summary text data DB 36, a keyword list is created for each summary text, and the keyword list 40 of each summary text is created. save.

  In step S104, a lyrics topic model 44 is created from each of the lyrics data stored in the lyrics data DB 30.

  In step S106, the lyrics topic is extracted from each of the lyrics data stored in the lyrics data DB 30 based on the lyrics topic model 44, and a topic list 48 of each lyrics is created.

  In step S108, the summary text topic model 52 for the summary text data is created by performing the same processing as in step S104 on the summary text data stored in the summary text data DB 36.

  In step S110, a summary text topic is extracted from each summary text data stored in the summary text data DB 36 based on the summary text topic model 52, and a topic list 56 of each summary text is created.

  In step S112, using the lyrics / outline text pair correct data 58, the lyrics that are a set of pairs in which the lyrics data of the music and the summary text data attached to the video not suitable for the lyrics data of the music are associated with each other. Outline text pair incorrect answer data 62 is created.

  In step S114, the keyword pair feature amount representing the combination of the lyric keyword extracted from the lyric data of the correct answer data and the outline text keyword extracted from the outline text data of the correct answer data and the lyric data of the incorrect answer data are extracted. From the keyword pair feature amount representing a combination of keywords for the summary text extracted from the summary text data of the lyric keywords and the incorrect answer data, and from the lyric topics and the summary text data of the correct answer data extracted from the lyrics data of the correct data Represents a combination of topic pair features that represent a combination of extracted summary text topics, a topic for lyrics extracted from lyrics data of incorrect answer data, and a topic for summary text extracted from summary text data of incorrect answer data. Topi Based on paired feature amount, it learns the ranking model, and stored in the ranking model storage unit 66, and ends the process.

  As described above, according to the learning device according to the embodiment of the present invention, the lyrics keyword is extracted from each of the lyrics data, the overview text keyword is extracted from each of the overview text data, and the lyrics data The topic for lyrics is extracted from each of the above, the topic for summary text is extracted from each of the summary text data, the keywords for lyrics extracted from the lyrics data of the correct answer data, and the overview extracted from the summary text data of the correct answer data Keyword pair feature amount representing a combination of text keywords and keyword pair feature amount representing a combination of a keyword for lyrics extracted from lyrics data of incorrect answer data and a keyword for summary text extracted from summary text data of incorrect answer data And a topic for lyrics extracted from the lyrics data of the correct answer data. Extracted from the summary text data of the topic pair representing the combination of the topic for the summary text extracted from the summary text data of the correct answer data and the topic for lyrics extracted from the lyrics data of the incorrect answer data and the incorrect answer data By learning the ranking model based on the topic pair feature amount representing the combination of the topics for the summary text, it is possible to learn the ranking model for accurately searching the video suitable for the lyrics data of the music.

<Operation of Video Retrieval Device According to Embodiment of the Present Invention>

  Next, the operation of the video search apparatus 200 according to the embodiment of the present invention will be described. When the lyrics data of the music is received by the input unit 210, the video search device 200 executes a video search processing routine shown in FIG.

  First, in step S200, a lyric keyword is extracted from the lyric data of the music received by the input unit 210.

  Next, in step S202, the lyrics topic is extracted from the lyrics data of the music received by the input unit 210 based on the lyrics topic model 244.

  In step S204, a combination of the lyric keyword extracted by the lyric keyword extracting unit 232 and the general text keyword extracted from the general text data added to the video in each general keyword list 240 for each video. This represents a combination of a keyword pair feature amount to be expressed, a lyrics topic extracted by the lyrics topic extraction unit 246, and a summary text topic extracted from the summary text data added to the video in the topic list 256 of each summary text. Based on the topic pair feature amount and the ranking model stored in the ranking model storage unit 266, the outline text attached to the video suitable for the lyrics data of the input music is searched.

  In step S206, N videos associated with the top N summary texts are output from the ranking information of the summary text searched in step S204, and the process is terminated.

  As described above, according to the video search apparatus according to the embodiment of the present invention, the keyword for lyrics is extracted from the lyrics data of the input music, and the lyrics topic is extracted from the lyrics data of the input music. For each extracted video, a keyword pair feature amount representing a combination of the extracted lyric keyword and the general text keyword extracted from the general text data attached to the video, and the extracted lyric topic and video By searching for a video suitable for the lyric data of the input music based on the topic pair feature amount representing the combination of the topics for the general text extracted from the provided general text data and the ranking model, Video suitable for lyric data can be searched with high accuracy.

  Also, if an appropriate video can be searched from long and subjective sentences such as the lyrics of the music, it will lead to a system that can provide video according to the music, and the way of enjoying the music will be expanded.

  The present invention is not limited to the above-described embodiment, and various modifications and applications can be made without departing from the gist of the present invention.

  For example, in the embodiment described above, the learning device 100 includes the lyrics data DB 30, the keyword extraction unit 32 for lyrics, the keyword list 34 for each lyrics, the summary text data DB 36, the keyword extraction unit 38 for summary text, Keyword list 40 for each summary text, topic model creation unit 42 for lyrics, topic model 44 for lyrics, topic extraction unit 46 for lyrics, topic list 48 for each lyrics, topic model creation unit 50 for summary text, The topic model 52 for the summary text, the topic extraction unit 54 for the summary text, the topic list 56 of each summary text, the lyrics / summary text pair correct answer data 58, the incorrect answer data creation part 60, and the lyrics / summary text pair. Incorrect answer data 62, ranker learning unit 64, ranking model storage unit 6 It is configured to include a preparative, and keywords paired feature amount, had learned the ranking model based on the topic paired feature amount is not limited thereto. For example, the learning device 100 includes a lyrics data DB 30, a keyword extraction unit 32 for lyrics, a keyword list 34 for each lyrics, a summary text data DB 36, a keyword extraction unit 38 for summary text, and a keyword list 40 for each summary text. , Lyrics / summary text pair correct answer data 58, incorrect answer data creation section 60, lyrics / summary text pair incorrect answer data 62, ranker learning section 64, ranking model storage section 66, and keyword pair The ranking model may be learned based on the feature amount. In this case, with respect to the video search device 200, the calculation unit 220 includes a lyrics keyword extraction unit 232, a keyword list 240 for each summary text, a video search unit 264, and a ranking model storage unit 266. The video suitable for the lyrics data of the input music may be searched based on the keyword pair feature amount and the ranking model.

  In addition, the learning apparatus 100 includes a lyrics data DB 30, a summary text data DB 36, a lyrics topic model creation unit 42, a lyrics topic model 44, a lyrics topic extraction unit 46, a topic list 48 for each lyrics, Summary text topic model creation unit 50, summary text topic model 52, summary text topic extraction unit 54, topic list 56 for each summary text, lyrics / summary text pair correct answer data 58, and incorrect answer data creation Unit 60, lyrics / outline text pair incorrect answer data 62, ranker learning unit 64, and ranking model storage unit 66, and the ranking model may be learned based on the topic pair feature amount. . In this case, with respect to the video search apparatus 200, the calculation unit 220 includes a topic model for lyrics 244, a topic extraction unit for lyrics 246, a topic list 256 for each summary text, a video search unit 264, and a ranking model storage unit. 266, and a video suitable for the lyrics data of the input music may be searched based on the topic pair feature quantity and the ranking model.

20, 220 Calculation unit 32, 232 Lyrics keyword extraction unit 34 Keyword list 38 for each lyrics 38 Outline text keyword extraction unit 40, 240 Keyword list 42 for each summary text Lyric topic model creation unit 44, 244 Lyric topic model 46 246 Lyrics topic extraction unit 48 Topic list 50 for each lyric Topic text topic model creation unit 52 General text topic model 54 General text topic extraction unit 56, 256 Topic list for each general text 58 Lyric / summary text pair correct answer Data 60 Incorrect answer data creation section 62 Lyric / summary text pair incorrect answer data 64 Ranker learning section 66 Ranking model storage section 100 Learning apparatus 200 Video search apparatus 210 Input section 264 Video search section 266 Ranking model storage section 2 0 output section

Claims (10)

A learning device for learning a ranking model for searching for a video suitable for song lyrics data,
Correct data that is a pair of the lyrics data of the music and the summary text data given to the video suitable for the lyrics data of the music, and the lyrics data of the music and the video that is not suitable for the lyrics data of the music A keyword extractor for lyrics from each of the lyrics data included in the incorrect answer data that is a pair with the summary text data,
A summary text keyword extraction unit that extracts a summary text keyword from each of the summary text data included in the correct answer data and the incorrect answer data;
A keyword pair feature amount representing a combination of a keyword for lyrics extracted from the lyrics data of the correct answer data and a keyword for summary text extracted from the summary text data of the correct answer data, and the lyrics data of the incorrect answer data A ranker learning unit that learns the ranking model based on the extracted keyword for lyrics and the keyword pair feature amount that represents a combination of keywords for summary text extracted from the summary text data of the incorrect answer data;
A learning device including A learning device for learning a ranking model for searching for a video suitable for song lyrics data,
Correct data that is a pair of the lyrics data of the music and the summary text data given to the video suitable for the lyrics data of the music, and the lyrics data of the music and the video that is not suitable for the lyrics data of the music A lyric topic extraction unit that extracts a lyric topic from each of the lyric data included in the incorrect answer data that is a pair with the summary text data;
A summary text topic extraction unit that extracts a summary text topic from each of the summary text data included in the correct answer data and the incorrect answer data;
A topic pair feature amount representing a combination of a topic for lyrics extracted from the lyrics data of the correct answer data and a topic for overview text extracted from the summary text data of the correct answer data, and the lyrics data of the incorrect answer data A ranker learning unit that learns the ranking model based on the extracted topic for lyrics and the topic pair feature amount representing the combination of the topic for summary text extracted from the summary text data of the incorrect answer data;
A learning device including A learning device for learning a ranking model for searching for a video suitable for song lyrics data,
Correct data that is a pair of the lyrics data of the music and the summary text data given to the video suitable for the lyrics data of the music, and the lyrics data of the music and the video that is not suitable for the lyrics data of the music A keyword extractor for lyrics from each of the lyrics data included in the incorrect answer data that is a pair with the summary text data,
A summary text keyword extraction unit that extracts a summary text keyword from each of the summary text data included in the correct answer data and the incorrect answer data;
A lyric topic extracting unit that extracts a lyric topic from each of the lyric data included in the correct answer data and the incorrect answer data;
A summary text topic extraction unit that extracts a summary text topic from each of the summary text data included in the correct answer data and the incorrect answer data;
A keyword pair feature amount representing a combination of a keyword for lyrics extracted from the lyrics data of the correct answer data and a keyword for summary text extracted from the summary text data of the correct answer data, and the lyrics data of the incorrect answer data A keyword pair feature amount representing a combination of an extracted text keyword and a summary text keyword extracted from the summary text data of the incorrect answer data, a lyrics topic extracted from the lyrics data of the correct data, and the Topic pair feature amount representing a combination of topics for summary text extracted from the summary text data of correct answer data, a topic for lyrics extracted from the lyrics data of the incorrect answer data, and the summary text data of the incorrect answer data Summary extracted from And Lanka learning portion based on the topic paired feature value representing the combination of text for topics, learns the ranking model,
A learning device including A video search device for searching video suitable for lyrics data of music,
A keyword extraction unit for lyrics that extracts a keyword for lyrics from the lyrics data of the input music;
From the lyric keyword extracted from the lyric data of the correct answer data that is a pair of the lyric data of the tune and the outline text data given to the video suitable for the lyric data of the tune, and the summary text data of the correct answer data The incorrect answer data that is a pair of the keyword pair feature amount representing the extracted summary text keyword combination, the song lyrics data, and the summary text data attached to the video not suitable for the song lyrics data Suitable for lyric data of music that has been learned in advance based on keyword pairs that represent combinations of keywords for lyrics extracted from lyrics data and keywords for summary text extracted from the summary text data of the incorrect answer data Ranking model that stores ranking models for searching for videos A storage unit,
A keyword pair feature amount representing a combination of a lyric keyword extracted by the lyric keyword extracting unit and an outline text keyword extracted from the outline text data attached to the image for each of the videos, and the ranking model Based on the above, a video search unit for searching for a video suitable for the lyrics data of the input music,
Video search device including A video search device for searching video suitable for lyrics data of music,
A lyric topic extractor that extracts lyric topics from the lyrics data of the input music;
From the lyrics topic extracted from the lyrics data of the correct answer data that is a pair of the lyrics data of the music and the outline text data assigned to the video suitable for the lyrics data of the music, and the summary text data of the correct answer data The incorrect answer data that is a pair of the feature pair feature amount representing the extracted summary text topic combination, the lyrics data of the music, and the summary text data attached to the video not suitable for the lyrics data of the music Suitable for lyric data of music that has been learned in advance based on the topic pair feature amount representing the combination of the topic for lyrics extracted from the lyric data and the topic for summary text extracted from the summary text data of the incorrect answer data A ranking model storage unit for storing a ranking model for searching for a captured video;
A topic pair feature amount representing a combination of a lyric topic extracted by the lyric topic extraction unit and an outline text topic extracted from the outline text data attached to the image for each of the videos, and the ranking model Based on the above, a video search unit for searching for a video suitable for the lyrics data of the input music,
Video search device including A video search device for searching video suitable for lyrics data of music,
A keyword extraction unit for lyrics that extracts a keyword for lyrics from the lyrics data of the input music;
A lyrics topic extraction unit that extracts a lyrics topic from the lyrics data of the input music;
From the lyric keyword extracted from the lyric data of the correct answer data that is a pair of the lyric data of the tune and the outline text data given to the video suitable for the lyric data of the tune, and the summary text data of the correct answer data The incorrect answer data that is a pair of the keyword pair feature amount representing the extracted summary text keyword combination, the song lyrics data, and the summary text data attached to the video not suitable for the song lyrics data A keyword pair feature amount representing a combination of a keyword for lyrics extracted from lyrics data and a keyword for summary text extracted from the summary text data of the incorrect answer data, and for lyrics extracted from the lyrics data of the correct answer data Extract from topic text data of topic and correct answer data A topic pair feature amount representing a combination of the extracted summary text topics, a lyrics topic extracted from the lyrics data of the incorrect answer data, and a summary text topic extracted from the summary text data of the incorrect answer data. A ranking model storage unit that stores a ranking model for searching for videos suitable for the lyric data of music, which has been learned in advance based on topic pair feature amounts representing combinations;
A keyword pair feature amount representing a combination of a lyric keyword extracted by the lyric keyword extracting unit and an outline text keyword extracted from the outline text data attached to the image for each of the videos, and for the lyrics Music composition inputted based on the topic pair feature amount representing the combination of the topic for lyrics extracted by the topic extraction unit and the topic for summary text extracted from the summary text data attached to the video, and the ranking model A video search unit to search for videos suitable for the lyrics data of
Video search device including   The learning device according to claim 1, wherein the summary text keyword is a keyword representing a person, a place, a season, or an event. A learning method in a learning device for learning a ranking model for searching for a video suitable for lyrics data of a song,
The lyric keyword extraction unit includes the correct data that is a pair of the lyric data of the music and the summary text data attached to the video suitable for the lyric data of the music, the lyric data of the music, and the lyric data of the music Extracting a lyric keyword from each of lyric data included in incorrect answer data that is a pair with summary text data attached to an unsuitable video;
A step of extracting a summary text keyword from each of the summary text data included in the correct answer data and the incorrect answer data;
A ranker learning unit includes a keyword pair feature amount representing a combination of a keyword for lyrics extracted from the lyrics data of the correct answer data and a keyword for summary text extracted from the summary text data of the correct answer data, and the incorrect answer data Learning the ranking model based on a keyword pair feature amount representing a combination of a lyric keyword extracted from the lyric data and a summary text keyword extracted from the summary text data of the incorrect answer data; ,
Learning methods including. A keyword extractor for lyrics, a keyword for lyrics extracted from the lyrics data of incorrect answer data which is a pair of the lyrics data of the song and the summary text data attached to the video not suitable for the lyrics data of the song; A ranking model for searching for a video suitable for lyric data of a song, learned in advance based on a keyword pair feature amount representing a combination of keywords for summary text extracted from the summary text data of the incorrect answer data A video search method in a video search device that includes a ranking model storage unit for storing, and a video search unit, and searches for a video suitable for lyrics data of a song,
The lyric keyword extraction unit extracts lyric keywords from the lyrics data of the input music;
A keyword pair feature representing a combination of a lyric keyword extracted by the lyric keyword extracting unit and a summary text keyword extracted from the summary text data attached to the video for each of the videos by the video search unit The keyword for lyrics extracted from the lyrics data of the correct answer data which is a pair of the amount, the lyrics data of the song, and the summary text data attached to the video suitable for the lyrics data of the song, and the summary of the correct data Search for a video suitable for the lyrics data of the input music based on the keyword pair feature amount representing the combination of keywords for the summary text extracted from the text data and the ranking model stored in the ranking model storage unit. And steps to
Video search method including   To make a computer function as each part of the learning device according to any one of claims 1 to 3 and claim 7 or the video search device according to any one of claims 4 to 6. Program.